- Use Blob instead of File for input in
ParquetFile.fromFile
. - Export wasm paths from
package.json
.
- Class-based API + concurrent streams + column selections + File reader by @H-Plus-Time in #407. This added a new
ParquetFile
API for working with files at remote URLs without downloading them first. - Conditional exports in
package.json
. This should make it easier to use across Node and browser. - Improved documentation for how to use different entry points.
- arrow2 and parquet2-based implementation has been removed.
- Layout of files has changed. Your import may need to change.
- Imports are now
parquet-wasm
,parquet-wasm/esm
,parquet-wasm/bundler
, andparquet-wasm/node
.
- Add conditional exports by @kylebarron in #382
- CI production build size summary by @H-Plus-Time in #401
- Remove arrow2 implementation by @kylebarron in #446
- feat: add lz4_raw support for
arrow1
by @fspoettel in #466 - Highlight that esm entry point needs await of default export by @kylebarron in #487
- Fixes for both report builds and PR comment workflow by @H-Plus-Time in #495
- fix package exports by @kylebarron in #414
- Object store wasm usage by @H-Plus-Time in #490
- Set Parquet key-value metadata by @kylebarron in #503
- Read parquet with options by @kylebarron in #506
- Documentation updates for 0.6 by @kylebarron in #507
- Avoid bigint for metadata queries by @kylebarron in #508
- Update async API by @kylebarron in #510
- Add test to read empty file by @kylebarron in #512
- bump arrow libraries to version 51 by @jdoig in #496
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.5.0...v0.6.0
- Switch to an API based on table abstractions from arrow-wasm.
- Update docs
- Initial implementation of reading to a stream of Arrow batches. By @H-Plus-Time in #296
- @H-Plus-Time made their first contribution in #296
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.4.0...v0.5.0
- Async reader support in the arrow2 bindings
- Improved memory usage via
readParquetFFI
in conjunction witharrow-js-ffi
. - Remove "2" from function names in arrow2 api by @kylebarron in #173
- Make arrow2 the default bundle by @kylebarron in #174
- Add bindings for arrow2 metadata (without serde support) by @kylebarron in #153
- Add lz4_raw and zstd compressions for parquet2 by @kylebarron in #114
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.3.1...v0.4.0
- Add lz4_raw and zstd compressions for parquet2 by @kylebarron in #114
- Simplify cargo features by @kylebarron in #117
- Add vscode rust-analyzer target setting by @kylebarron in #131
- add msrv by @kylebarron in #132
- pin clap to 3.1.* by @kylebarron in #139
- Make writerProperties optional in JS api by @kylebarron in #152
- Add bindings for arrow2 metadata (without serde support) by @kylebarron in #153
- Async reader by @kylebarron in #96
- Cleaner error handling by @kylebarron in #157
- implement
From
instead of custom methods by @kylebarron in #168 - Remove "2" from function names in arrow2 api by @kylebarron in #173
- Make arrow2 the default bundle by @kylebarron in #174
- Improved documentation for async reading by @kylebarron in #175
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.3.1...v0.4.0-beta.1
- Bump arrow from 11.0.0 to 11.1.0 by @dependabot in #77
- Update lockfile by @kylebarron in #76
- Add clippy by @kylebarron in #78
- Remove old debug script by @kylebarron in #79
- Bump clap from 3.1.8 to 3.1.9 by @dependabot in #87
- Check that input exists/is a uint8array by @kylebarron in #102
- Update test files to those written by pyarrow v7 by @kylebarron in #103
- Update to arrow and parquet 12.0 by @kylebarron in #105
- Bump clap from 3.1.9 to 3.1.12 by @dependabot in #98
- Create arrow1/arrow2 read benchmarks by @kylebarron in #82
- Publish docs on tag by @kylebarron in #106
- Update readme by @kylebarron in #107
- Add published examples section to readme by @kylebarron in #108
- Unify build script by @kylebarron in #109
- esm2 entrypoint with no import.meta.url by @kylebarron in #110
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.3.0...v0.3.1
- Debug cli by @kylebarron in #64
- Bump to arrow 11.0 to support zstd compression by @kylebarron in #66
- Update bundling by @kylebarron in #67
- Add dependabot by @kylebarron in #70
- Bump clap from 3.1.6 to 3.1.8 by @dependabot in #71
- Bump getrandom from 0.2.5 to 0.2.6 by @dependabot in #72
- @dependabot made their first contribution in #71
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.2.0...v0.3.0
- Restore arrow-rs support by @kylebarron in #21
- Write parquet with arrow1 by @kylebarron in #23
- Refactor code into lower-level functions, use
?
operator by @kylebarron in #25 - Make record batch size the nrows of the first row group by @kylebarron in #26
- Rename arrow-rs api as default by @kylebarron in #31
- Implement writerPropertiesBuilder for arrow1 by @kylebarron in #30
- Refactor into modules by @kylebarron in #32
- Update bundling to create arrow2 entrypoints by @kylebarron in #33
- Node testing setup by @kylebarron in #34
- Helper to copy vec to Uint8Array by @kylebarron in #38
- Faster builds on Node CI tests by @kylebarron in #39
- Rust CI caching by @kylebarron in #40
- ZSTD mac instructions in readme by @kylebarron in #42
- Keep opt-level = s and remove
console_error_panic_hook
by @kylebarron in #48 - WriterPropertiesBuilder for arrow2 by @kylebarron in #49
- Docstrings for public functions, structs, enums by @kylebarron in #50
- Compression-specific features by @kylebarron in #51
- Add more node tests by @kylebarron in #52
- Separate reader and writer features by @kylebarron in #47
- Docs update by @kylebarron in #53
- Working typedoc by @kylebarron in #55
- Update docstrings and readme by @kylebarron in #60
Full Changelog: https://github.com/kylebarron/parquet-wasm/compare/v0.1.1...v0.2.0
- Attempt better bundling, with APIs for bundlers, Node, and the Web.
- Initial release
- Barebones
read_parquet
andwrite_parquet
functions.